seedance-camera

Pass

Audited by Gen Agent Trust Hub on Apr 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill contains only documentation, framing presets, and usage examples without any executable scripts, binary files, or dangerous system commands.
  • [PROMPT_INJECTION]: The documentation identifies a surface for indirect prompt injection through the use of the seedance command which processes external media.
  • Ingestion points: The command examples in SKILL.md use --image and --video flags to ingest untrusted media files.
  • Boundary markers: There are no specified delimiters or instructions to ignore embedded prompts within the provided media files in the examples.
  • Capability inventory: The documented seedance CLI tool is used for automated video generation and editing.
  • Sanitization: No mention of media content validation or prompt sanitization is provided in the instructional text.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 23, 2026, 10:08 AM